# llms.txt

# This is a hypothetical llms.txt file, mirroring your robots.txt directives.
# IMPORTANT: There is no official standard for llms.txt recognized by all LLM providers,
# and this file cannot be directly uploaded to your Shopify store without workarounds.
# It serves as an illustrative example of how such directives might be structured.

# General LLM bot rules (equivalent to User-agent: * in robots.txt)
User-agent: *LLM* # Represents any Large Language Model crawler
Disallow: /admin
Disallow: /cart
Disallow: /checkout
Disallow: /checkouts/
Disallow: /orders
Disallow: /account
Disallow: /carts
Disallow: /*?oseid=
Disallow: /*preview_theme_id
Disallow: /*preview_script_id
Disallow: /apple-app-site-association
Disallow: /.well-known/shopify/monorail
Disallow: /cdn/wpm/*.js
Disallow: /recommendations/products

# These are allowed for LLM indexing (DO NOT BLOCK):
Allow: /collections/
Allow: /products/
Allow: /blogs/
Allow: /pages/
Allow: /search # incl. sorted, filtered, and tagged content

# Google-specific LLM directive
User-agent: Google-Extended # Official bot used by Google for Gemini/Bard models
Disallow: /admin/
Allow: /

# OpenAI's GPT-based models (ChatGPT, etc.)
User-agent: GPTBot
Disallow: /admin/
Allow: /

# Hypothetical LLM bot for advertising content
User-agent: AdLLMBot
Disallow: /checkout
Disallow: /checkouts/
Disallow: /carts
Disallow: /orders

# Hypothetical LLM bot for image processing (e.g., future visual model indexing)
User-agent: ImageLLMBot
Disallow: /admin/
Allow: /

# Hypothetical LLM bot for Pinterest
User-agent: PinterestLLM
Crawl-delay: 1

# Hypothetical LLM bots for Ahrefs
User-agent: AhrefsLLMBot
Crawl-delay: 10
Disallow: /checkout
Disallow: /checkouts/
Disallow: /cart
Disallow: /admin
Disallow: /orders

User-agent: AhrefsSiteAuditLLM
Crawl-delay: 10
Disallow: /checkout
Disallow: /checkouts/
Disallow: /cart
Disallow: /admin
Disallow: /orders

# Hypothetical LLM bot for MJ12bot
User-agent: MJ12LLMBot
Crawl-delay: 10

# Block outdated or aggressive LLM scrapers
User-agent: OutdatedLLMScraper
Disallow: /

# Best Uses for a hypothetical llm.txt file:
# 1. Data Privacy & Security: Prevent LLM crawlers from accessing sensitive areas.
# 2. Content Control for AI Training: Explicitly allow/disallow content for AI model training.
# 3. Resource Management: Potentially mitigate server load by requesting crawl delays.
# 4. Intellectual Property Protection: Signal content usage preferences.
# 5. Targeted Directives: Apply specific rules to different LLM user agents.